{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,13]],"date-time":"2026-05-13T16:03:58Z","timestamp":1778688238024,"version":"3.51.4"},"reference-count":13,"publisher":"SAGE Publications","issue":"3","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IFS"],"published-print":{"date-parts":[[2021,3,2]]},"abstract":"<jats:p>Generalization ability is known as an important performance index of artificial neural networks (ANNs). The generalization ability of an ANN usually refers to its ability to recognize untrained samples, but it lacks quantitative analysis. A method is designed by using frequency-domain signals to observe the generalization ability of deep feedforward neural networks (DFFNNs) which are popular ANN models. This method allows us to observe that the generalization ability of DFFNNs is limited to a small neighborhood around the trained samples. Then, the relationship between sample similarity and the DFFNN\u2019s generalization performance is further analyzed. The analysis results show that the correlation coefficient between samples has a certain positive correlation with the DFFNN\u2019s generalization performance. Based on this new understanding, an algorithm in which shadows of the trained samples are added into the training set is proposed to improve the generalization ability of DFFNNs. The proposed algorithm is tested with some simulated signals and some real-world data. The tests show that the proposed method can indeed improve the DFFNN\u2019s generalization ability by only changing the training sample set.<\/jats:p>","DOI":"10.3233\/jifs-201679","type":"journal-article","created":{"date-parts":[[2021,1,26]],"date-time":"2021-01-26T12:43:23Z","timestamp":1611665003000},"page":"4867-4876","source":"Crossref","is-referenced-by-count":12,"title":["Quantitative analysis of the generalization ability of deep feedforward neural networks"],"prefix":"10.1177","volume":"40","author":[{"given":"Yanli","family":"Yang","sequence":"first","affiliation":[{"name":"Tianjin Key Laboratory of Optoelectronic Detection Technology and Systems, Tiangong University, Tianjin, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Chenxia","family":"Li","sequence":"additional","affiliation":[{"name":"Tianjin Key Laboratory of Optoelectronic Detection Technology and Systems, Tiangong University, Tianjin, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"179","reference":[{"key":"10.3233\/JIFS-201679_ref1","doi-asserted-by":"crossref","first-page":"1638","DOI":"10.1109\/TFUZZ.2014.2371479","article-title":"A study on relationship between generalization abilities and fuzziness of base classifiers in ensemble learning","volume":"23","author":"Wang","year":"2015","journal-title":"IEEE Transactions on Fuzzy Systems"},{"key":"10.3233\/JIFS-201679_ref3","doi-asserted-by":"crossref","first-page":"9241","DOI":"10.1007\/s00521-019-04129-5","article-title":"Evaluating generalization through interval-based neural network inversion,","volume":"31","author":"Adam","year":"2019","journal-title":"Neural Computing and Applications"},{"key":"10.3233\/JIFS-201679_ref8","doi-asserted-by":"crossref","first-page":"145","DOI":"10.1109\/JSAIT.2020.2983192","article-title":"A Fourier-based approach to generalization and optimization in deep learning,","volume":"1","author":"Farnia","year":"2020","journal-title":"IEEE Journal on Selected Areas in Information Theory"},{"key":"10.3233\/JIFS-201679_ref11","doi-asserted-by":"crossref","first-page":"85","DOI":"10.1016\/j.neunet.2020.06.024","article-title":"Quantifying the generalization error in deep learning in terms of data distribution and neural network smoothness,","volume":"130","author":"Jin","year":"2020","journal-title":"Neural Networks"},{"key":"10.3233\/JIFS-201679_ref12","doi-asserted-by":"crossref","first-page":"85","DOI":"10.1016\/j.neunet.2014.09.003","article-title":"Deep learning in neural networks: An overview,","volume":"61","author":"Schmidhuber","year":"2015","journal-title":"Neural Networks"},{"key":"10.3233\/JIFS-201679_ref13","doi-asserted-by":"crossref","unstructured":"Yang Y.L. and He Y.C. , Aweight initialization method associated with samples for deep feedforward neural network, In: International Conference on Artificial Intelligence in Electronics Engineering, Sanya, China, 2020.","DOI":"10.1145\/3379247.3379253"},{"key":"10.3233\/JIFS-201679_ref14","first-page":"504","article-title":"Reducing the dimensionality of data with","volume":"313","author":"Hinton","year":"2006","journal-title":"Neural Networks Science"},{"key":"10.3233\/JIFS-201679_ref17","doi-asserted-by":"crossref","first-page":"71540","DOI":"10.1109\/ACCESS.2018.2880990","article-title":"Bearing faults automatic classification based on deep learning,","volume":"6","author":"Yang","year":"2018","journal-title":"IEEE Access"},{"key":"10.3233\/JIFS-201679_ref19","doi-asserted-by":"crossref","first-page":"59","DOI":"10.2307\/2685263","article-title":"Thirteen ways to look at the correlation coefficient,","volume":"42","author":"Rodgers","year":"2018","journal-title":"The American Statistician"},{"key":"10.3233\/JIFS-201679_ref20","doi-asserted-by":"crossref","first-page":"35","DOI":"10.1177\/875647939000600106","article-title":"Interpretation of the correlation coefficient: a basic review,","volume":"6","author":"Taylor","year":"1990","journal-title":"Journal of Diagnostic Medical Sonography"},{"key":"10.3233\/JIFS-201679_ref25","doi-asserted-by":"crossref","first-page":"1745","DOI":"10.1016\/j.dib.2018.11.019","article-title":"Bearing vibration data collected under time-varying rotational speed conditions","volume":"21","author":"Huang","year":"2018","journal-title":"Data in Brief"},{"key":"10.3233\/JIFS-201679_ref26","doi-asserted-by":"crossref","first-page":"401","DOI":"10.1109\/TR.2018.2882682","article-title":"Ahybrid prognostics approach for estimating remaining useful life of rolling element bearings","volume":"69","author":"Wang","year":"2020","journal-title":"IEEE Transactions on Reliability"},{"key":"10.3233\/JIFS-201679_ref27","doi-asserted-by":"crossref","first-page":"100","DOI":"10.1016\/j.ymssp.2015.04.021","article-title":"Randall, Rolling element bearing diagnostics using the Case Western Reserve University data: A benchmark study","volume":"64\u201365","author":"Smith","year":"2015","journal-title":"Mechanical Systems and Signal Processing"}],"container-title":["Journal of Intelligent &amp; Fuzzy Systems"],"original-title":[],"link":[{"URL":"https:\/\/content.iospress.com\/download?id=10.3233\/JIFS-201679","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T09:43:28Z","timestamp":1777455808000},"score":1,"resource":{"primary":{"URL":"https:\/\/journals.sagepub.com\/doi\/full\/10.3233\/JIFS-201679"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,3,2]]},"references-count":13,"journal-issue":{"issue":"3"},"URL":"https:\/\/doi.org\/10.3233\/jifs-201679","relation":{},"ISSN":["1064-1246","1875-8967"],"issn-type":[{"value":"1064-1246","type":"print"},{"value":"1875-8967","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,3,2]]}}}